UNIVERSITÀ DEGLI STUDI DI TRIESTE
XXXI
CICLO
DEL
DOTTORATO
DI
RICERCA
IN
NANOTECNOLOGIE
BIORESORBABLE ENGINEERED MEMBRANES FOR GUIDED
BONE REGENERATION WITH ANTIMICROBIAL PROPERTIES
Settore scientifico disciplinare: MED/28
DOTTORANDO
Federico Berton
COORDINATORE
Chiar.ma Prof. Lucia Pasquato
SUPERVISORE DI TESI
Chiar.mo Prof. Roberto Di Lenarda
CO-SUPERVISORI DI TESI
Prof. Gianluca Turco
Dott. Davide Porrelli
TABLE OF CONTENTS
LIST OF THE ABBREVIATIONS ... 8
ABSTRACT ... 10
RIASSUNTO... 14
INTRODUCTION AND THESIS OBJECTIVES ... 18
GENERAL INTRODUCTION ... 18
BONE REGENERATION VIA ELECTROSPUN NANOFIBER MEMBRANES ... 19
BONE REGENERATION WITH ENRICHED HEMODERIVATIVES ... 22
BONE REGENERATION ANALYSIS ... 23
AIM OF THE RESEARCH ... 25
CHAPTER 1 ELECTROSPINNING IN BONE REGENERATION ... 26
VARIABLES AFFECTING ELECTROSPINNING PROCESS ... 26
MATERIAL AND METHODS ... 30
ELECTROSPINNING DEVICE ASSEMBLY ... 31
EXPERIMENTAL SETUP ... 32
MATERIALS ... 32
PREPARATION OF PCL SOLUTIONS ... 33
SOLVENTS AND PARAMETERS OF PCL SOLUTIONS FOR ELS ... 33
SCANNING ELECTRON MICROSCOPE (SEM) ... 35
PLASMA-AIR TREATMENT ... 35
CONTACT ANGLE ... 35
CTL PRODUCTION ... 36
CTL NAG PRODUCTION ... 36
COMPUTERIZED MICRO-TOMOGRAPHY (ΜCT) ... 36
CTL CONFOCAL MICROSCOPE ANALYSIS OF ADSORPTION ON ELECTROSPUN PCL MEMBRANES ... 36
FTIR ANALYSIS... 37
RAMAN ANALYSIS ... 37
ELECTROTHERMAL ATOMIC ABSORPTION SPECTROMETRY (ETAAS) ... 37
SIMULATED BODY FLUID (SBF) PRODUCTION AND MEMBRANE DISSOLUTION ASSAY ... 38
MECHANICAL TESTING WITH DMA ... 38
CELL CULTURE AND IN VITRO TESTS ... 38
SEM AND CELL ADHESION AND PROLIFERATION TESTS ... 39
CELL PROLIFERATION ... 39 CELL TOXICITY ... 40 BIOFILM INHIBITION ... 40 RESULTS... 42 PCL-BASED SOLUTIONS ... 42 ΜCT EXAMINATION OF PCL MEMBRANES ... 45
MECHANICAL TESTING STABILITY OF ELECTROSPUN MEMBRANES ... 47
CTL E CTL-NAG ADSORPTION ON PCL MEMBRANES ... 50
CONFOCAL MICROSCOPE ASSESSMENT OF CTL ADSORPTION ... 53
FT-IR ANALYSIS ... 54
RAMAN ANALYSIS ... 55
ELECTROTHERMAL ATOMIC ABSORPTION SPECTROMETRY (ETAAS) ... 56
PCL,PCL-CTL AND PCL-CTL-NAG MEMBRANE WETTABILITY TEST THROUGH CONTACT ANGLE MEASUREMENTS ... 57
EFFECT OF PLASMA AIR TREATMENT ON CONTACT ANGLE EXAMINATION ... 57
ALAMAR BLUETM ASSAY ... 63
CYTOTOXICITY TEST WITH LDH ... 65
SEM ANALYSIS OF HUMAN MG63 OSTEOSARCOMA CELLS GROWN ON ELECTROSPUN MEMBRANES ... 66
CHAPTER 2 AUTOLOGOUS NANOFIBER-BASED MEMBRANES IN BONE REGENERATION: PLATELET RICH FIBRIN ... 68
MATERIAL AND METHODS ... 69
MATERIALS ... 69
MEMBRANE PRODUCTION ... 69
ALGINATE-HAP FREEZE DRIED SCAFFOLD PRODUCTION ... 70
HAP ENRICHMENT: FIRST EXPERIMENT ... 71
HAP ENRICHMENT: SECOND EXPERIMENT ... 71
SEM MORPHOLOGIC ANALYSIS ... 72
SEMEDS CHEMICAL ANALYSIS ... 73
MECHANICAL TESTING WITH DMA ... 73
SIMULATED BODY FLUID (SBF) PRODUCTION ... 73
SBF DISSOLUTION ASSAY ... 73
RESULTS... 74
SEM MORPHOLOGIC ANALYSIS OF PRISTINE PRF MEMBRANES ... 74
SBF DISSOLUTION ASSAY ... 75
MECHANICAL TESTING WITH DMA ... 81
NHAP ENRICHMENT FIRST EXPERIMENT ... 82
SEMEDS MORPHOLOGIC AND CHEMICAL ANALYSIS OF HAP-ENRICHED PRF MEMBRANES SECOND EXPERIMENT ... 84
CHAPTER 3 BONE REGENERATION ANALYSIS ... 88
MATERIAL AND METHODS ... 89
MATERIALS ... 89
SAMPLES PREPARATION: COLLECTION ... 90
COLLECTION OF SAMPLE: CASE 1... 90
COLLECTION OF SAMPLE: CASE 2... 90
COLLECTION OF SAMPLE: CASE 3... 91
COLLECTION OF SAMPLE: CASE 4... 91
SAMPLES PREPARATION: RESIN INCLUSION... 91
SAMPLES PREPARATION: CUTTING ... 92
SAMPLES PREPARATION: LAPPING ... 92
SAMPLES PREPARATION: POLISHING ... 92
SAMPLES STAINING: TOLUIDINE BLUE AND ACID FUCHSINE ... 92
SAMPLES STAINING:VON KOSSA AND RED NEUTRAL... 93
SAMPLES STAINING:MASSON TRICHROME ... 93
SAMPLES ANALYSIS: OPTICAL MICROSCOPE IMAGE ACQUISITION AND MERGING ... 93
HISTOMORPHOMETRIC INDICES ... 95
SAMPLES PREPARATION FOR SEM-EDS ANALYSIS ... 96
SEM-EDS ANALYSIS ... 96
RESULTS... 97
CASE 1:HISTOMORPHOMETRIC ANALYSIS OF HOMOLOGOUS BONE BLOCK ALTER 9 YEARS VON KOSSA STAINING ... 97
CASE 2:HISTOMORPHOMETRIC AND SEM-EDS ANALYSIS OF EX VIVO RETRIEVED TITANIUM IMPLANT FROM BOVINE BONE .. 100
CASE 3:HISTOMORPHOMETRIC ANALYSIS OF HUMAN BONE REGENERATION, TOLUIDINE BLUE AND ACID FUCHSINE STAIN ... 104
ELECTROSPUN NANOFIBER MEMBRANES IN BONE REGENERATION ... 108
BONE REGENERATION WITH ENRICHED HEMODERIVATIVES ... 111
BONE REGENERATION ANALYSIS ... 112
CONCLUSION AND FUTURE PLANS ... 114
List of the abbreviations
β-TCP beta tricalcium phosphate BIC bone to implant contact BMPs bone morphogenetic proteins BPCi conductivity index of bone particles BV/TV bone volume/total volume
CTL lactose-modified chitosan
CTH chitosan
DBBM demineralized bovine bone matrix DCM dichloromethane
DMA dynamic mechanical analysis DMEM Dulbecco's modified eagle medium DMF dimethylformamide
DW distilled water
EDS or EDX energy-dispersive X-ray spectroscopy ELS electrospinning
ETAAS electrothermal atomic absorption spectrometry FITC fluorescein isothiocyanate
GBR guided bone regeneration GFs growth factors
GDL δ-glucono lactone
GPa gigapascal
HMDS hexamethyldisilazane
ICP-MS inductively coupled plasma mass spectrometry IGF-1 insulin like growth factor 1
L-PRF leukocyte – platelet rich fibrin
LB Luria Bertani
LDH lactate dehydrogenase LWR London white resin
μCT micro computed tomography MTT thiazolyl blue tetrazolium bromide MW molecular weight
nAg silver nanoparticles
NB native bone
NFB newly formed bone nHAp nanohydroxyapatite PBS Phosphate-buffered saline PCL polycaprolactone
PCs platelet concentrates
PDGF platelet derived growth factor PEO polyethylene oxide
PRF platelet rich fibrin PRP platelet rich plasma
Ra roughness average
RG residual graft
SBF simulated body fluid
TFA trifluoroacetic acid Tb.N trabecular number Tb.Sp trabecular spacing Tb.Th trabecular thickness
TGF-β transforming growth factor beta THF tetrahydrofuran
TrSD trabecular space dimensions UniTS University of Trieste
Abstract
Background
The main topics of this doctoral thesis are bone regeneration and bone tissue analysis by means of nanotechnological strategies for both purposes. Briefly, the development of nanostructured membranes and scaffolds for bone regeneration procedures in oral surgery have been explored, together with the optimization of analysis protocols at the micro- and nano-scale. Therefore, three main chapters will be here described: (i) nanostructured electrospun membranes production, tailoring and analysis; (ii) nanofiber fibrin-based membrane production, tailoring and analysis; (iii) development of sound and reliable histological, histomorphometric, chemical and ultrastructural protocols for the analysis of hard tissues.
Materials and Methods
Considered that, at the first stages of this PhD project, the electrospinning device was not available at the University of Trieste (UniTS) Labs, the efforts were firstly conducted to i) acquire the knowledge about this technique and ii) the application of this methodology to the synthesis of biocompatible membranes.
The first attempts were addressed to reproduce some of the recent results reported in the scientific literature. Subsequently, thanks to the collaboration with the Tissue Engineering Laboratory,
Campus Biomedico of Rome (Prof. Alberto Rainer), specific polymeric solutions have been prepared.
Different solutions were prepared with promising results in terms of proper solvent evaporation, homogeneity of fiber diameters, and absence of defects such as beads, when electrospun.
Thereafter the necessary components for the ELS process were acquired, assembled and tested at UniTS. Different combinations of solvents for the preparation of polycaprolactone (PCL) based membranes were tested. Thereafter, an air-plasma activation process was applied in order to increase the membrane hydrophilicity. Qualitative characterization of membranes obtained by ELS, before and after plasma treatment has been performed together with the morphological analysis (orientation, presence of beads, fiber diameter) through Scanning Electronic Microscopy. The determination of surface wettability with contact angle measurements was performed. The optimal formulation was chosen for subsequent evaluation of hydrophilicity and morphology following the activation of the membranes.
Spectrometry (ETAAS) analysis was performed to quantify nAg. MG63 cells cultured on PCL membranes with or without CTL were used for proliferation assay with daily timepoints until day eight. Lactate dehydrogenase (LDH) assay was used for cytotoxicity evaluation. The antibacterial activity of PCL-nAg membranes was tested in terms of biofilm inhibition on Pseudomonas
aeruginosa (ATCC 27853) and Staphylococcus aureus (ATCC 25923), using the MTT test. Mechanical
properties of the produced membranes were evaluated in dry conditions, after rapid soaking and after aging in Simulated Body Fluid (SBF). These tests were performed by means of uniaxial tensile tests for the evaluation of the elastic modulus, the deformation at failure and the ultimate tensile stress.
As second chapter of this thesis, the potentiality of another approach to produce nanostructured fibrin-based membranes was explored. These membranes were obtained by means of blood centrifugation and were tailored with a bone-inducing molecule, i.e. nano-hydroxyapatite (nHAp) (nanopowder, particle size <200 nm) for bone regeneration. The stability of pristine membrane in SBF was tested. At each time point, one sample was analyzed by means of SEM and subsequently with ImageJ processing tools. Then, the effects of the nHAp addition during the forming process of PRF (thus during centrifugation) were investigated.
Results
The production of nanostructured electrospun based bioactive membranes with antimicrobial properties started with the activation of pristine PCL membranes. The activation of the highly hydrophobic PCL membranes by means of the plasma-air treatment resulted in a drastic decrease of the contact angle. The resulted membranes had an estimated average thickness of 137.3 ± 7.0 μm and 215.6 ± 22.1 μm after 30 and 60 minutes of fiber deposition, respectively. Moreover, the membranes exhibited appropriate handling features for clinical use. In terms of mechanical properties, tested by means of uniaxial tensile test, plasma-activated PCL membranes exhibited higher toughness if compared to the untreated ones. Confocal microscopy analysis showed an improved adsorption of CTL (labelled with fluorescein) for the membranes treated with air-plasma if compared with the untreated ones. CTL adsorption was confirmed by means of ETAAS which showed a higher nAg content in membranes treated with a low energy air-plasma treatment and CTL-nAg at pH 7, thus confirming Raman findings. MG63 cells cultured on PCL membranes with or without CTL, showed a more sustained growth after 7 days on the CTL-coated membranes compared with untreated PCL membranes and PCL air-plasma treated membranes. Moreover, the presence of nAg did not hamper cell viability with respect to PCL membranes, as confirmed by LDH assay. The antibacterial activity of PCL-nAg clearly showed that the biofilm formation was strongly inhibited on the surface of PCL-CTL-nAg membranes. Mechanical resistance of the produced membranes, soaked and aged in Simulated Body Fluid (SBF) showed results superior to commercially available membranes.
For the experimental enrichment of PRF membranes the dissolution assay of the pristine membrane revealed a mean fiber diameter of 0.103 ± 0.05µm without any statistically significant differences during time; degradation assay showed a two-folds increase of the weight related to the SBF absorption in the first 2 days. From the third day a constant degradation was observed. In the time frame of this experiment, the dimensional stability of the fibrin structure up to day 7 suggested that PRF membranes may also be used uncovered in the oral cavity. After the two methods of tested nHAp enrichment for each condition a sediment of nHAp was observed to be present on the bottom of the vials. On the other hand, a limited amount of nHAp coating the fibers was detected by means of SEM-EDS analysis. Further efforts will be made to find out the best conditions to obtain a suitable coupling of the two agents without hampering the formation of fibers.
volume/total volume; regenerated bone; residual graft particles; soft tissue; conductivity index; trabecular thickness and trabecular spacing; bone-to-implant contact ratio (BIC). For the chemical and microscopic analysis of hard tissues, SEM-EDS was used for calcium and phosphate quantification in different regions of the bone sample, after the re-lapping of the histological slide and its carbon coating. A specimen containing a titanium implant was analyzed in order to evaluate its interface with bone, and the cortical bone and the cancellous bone grown around the implant. Even if reference values are, to date, missing in the literature, it was possible to detect different grades of mineralization at the interface (different calcium concentration) and between cancellous and cortical bone.
Conclusions
The membranes produced by means of ELS showed promising results in terms of (i) reliability of production (ii) biocompatibility (ii) antimicrobial properties (iii) stability in SBF allowing the scheduling of further in vivo experiments with the obtained membranes. Parallelly, the experimental enrichment of PRF membranes, even showing nHAp adsorption, still need to be perfected to obtain a homogenous distribution of nHAp on membrane’s surface.
Riassunto
Introduzione
Gli argomenti principali di questa Tesi di Dottorato sono la rigenerazione ossea e l'analisi del tessuto osseo mediante strategie nanotecnologiche per entrambi gli scopi. In breve, sono stati esplorati lo sviluppo di membrane e scaffold nanostrutturati per le procedure di rigenerazione ossea in chirurgia orale, insieme all'ottimizzazione dei protocolli di analisi su micro e nanoscala. Pertanto, verranno descritti tre capitoli principali: (i) produzione, ottimizzazione e analisi di membrane elettrofilate nanostrutturate; (ii) produzione, ottimizzazione e analisi di membrane a base di fibrina di nanofibre; (iii) sviluppo di protocolli istologici, istomorfometrici, chimici e molecolari riproducibili e affidabili per l'analisi dei tessuti duri.
Materiali e metodi
Tenendo conto del fatto che, nelle prime fasi di questo progetto di Dottorato, i dispositivi di
electrospinning non erano disponibili presso i Laboratori dell'Università degli Studi di Trieste, gli
sforzi sono stati inizialmente condotti per i) acquisire le conoscenze su questa tecnica e ii) l'applicazione di questa metodologia per la sintesi di membrane biocompatibili.
I primi tentativi furono indirizzati a riprodurre alcuni dei recenti risultati riportati nella letteratura scientifica. Successivamente, grazie alla collaborazione con il Tissue Engineering Laboratory, Campus Biomedico di Roma (Prof. Alberto Rainer), sono state preparate alcune soluzioni polimeriche. Sono state preparate diverse soluzioni con risultati promettenti in termini di corretta evaporazione del solvente, omogeneità dei diametri delle fibre e assenza di difetti come le beads. Successivamente sono stati acquisiti, assemblati e testati i componenti necessari per il processo di ELS. Sono state testate diverse combinazioni di solventi per la preparazione di membrane a base di policaprolattone (PCL). Successivamente, è stato applicato un processo di attivazione aria-plasma al fine di aumentare l'idrofilia della membrana. La caratterizzazione qualitativa delle membrane ottenute da ELS, prima e dopo il trattamento al plasma, è stata eseguita insieme all'analisi morfologica (orientamento, presenza di beads, diametro della fibra) attraverso la microscopia elettronica a scansione. È stata determinata la bagnabilità della superficie con misurazioni dell'angolo di contatto. La formulazione ottimale è stata scelta per la successiva valutazione di idrofilia e morfologia a seguito dell'attivazione delle membrane.
L'analisi ETAAS è stata eseguita per quantificare nAg. Le cellule MG63 coltivate su membrane PCL con o senza CTL sono state utilizzate per il saggio di proliferazione con timepoints giornalieri fino all'ottavo giorno. Il dosaggio della lattato deidrogenasi (LDH) è stato utilizzato per la valutazione della citotossicità. L'attività antibatterica delle membrane PCL-nAg è stata testata in termini di inibizione del biofilm su Pseudomonas aeruginosa (ATCC 27853) e Staphylococcus aureus (ATCC 25923), utilizzando il test MTT. La resistenza meccanica delle membrane prodotte è stata valutata dopo immersione e invecchiamento in Simulated Body Fluid (SBF), mediante prove di trazione uniassiali per la valutazione del modulo elastico, della deformazione a rottura e dello sforzo limite. Come secondo capitolo di questa tesi, è stata esplorata la potenzialità di un altro approccio per produrre membrane a base di fibrina nanostrutturate. Queste membrane sono state ottenute mediante centrifugazione del sangue e sono state implementate con nano-idrossiapatite (nHAp) (nanopolvere, particelle di dimensioni <200 nm) per la rigenerazione ossea. È stata testata la durabilità di membrane native in SBF. Ad ogni tempo di indagine stabilito, un campione è stato analizzato con SEM e successivamente con gli strumenti di elaborazione del programma ImageJ. Inoltre, sono stati studiati gli effetti dell'aggiunta di nHAp durante il processo di formazione di PRF (quindi durante la centrifugazione).
Risultati
La produzione di membrane bioattive nanostrutturate a base di PCL elettrofilate con proprietà antimicrobiche è iniziata con l'attivazione di membrane PCL native. L'attivazione delle membrane PCL altamente idrofobiche mediante il trattamento plasma-aria ha comportato una drastica riduzione dell'angolo di contatto. Le membrane risultanti avevano uno spessore medio stimato di 137,3 ± 7,0 μm e 215,6 ± 22,1 μm dopo 30 e 60 minuti di deposizione di fibre, rispettivamente. Inoltre, le membrane presentavano caratteristiche di maneggevolezza appropriate per l'uso clinico. In termini di proprietà meccaniche, testate mediante test di trazione uniassiale, le membrane PCL attivate al plasma hanno mostrato una maggiore tenacità rispetto a quelle non trattate. L'analisi microscopica confocale ha mostrato un miglior assorbimento di CTL (marcato con fluoresceina) per le membrane trattate con plasma ad aria rispetto a quelle non trattate. L'adsorbimento di CTL è stato confermato mediante ETAAS che ha mostrato un contenuto di nAg più elevato nelle membrane trattate con un trattamento plasma-aria a bassa energia e CTL-nAg a pH 7, confermando così i risultati delle analisi al Raman. Le cellule MG63 coltivate su membrane PCL con o senza CTL, hanno mostrato una crescita più sostenuta dopo 7 giorni sulle membrane rivestite CTL rispetto alle membrane PCL non trattate e alle membrane trattate con aria-plasma PCL. Inoltre, la presenza di nAg non ha ostacolato la vitalità cellulare rispetto alle membrane PCL, come confermato dal test LDH. L'attività antibatterica di PCL-nAg ha mostrato chiaramente che la formazione di biofilm era fortemente inibita sulla superficie delle membrane di PCL-CTL-nAg. La resistenza meccanica delle membrane prodotte, imbevute e invecchiate nel fluido corporeo simulato (SBF) ha mostrato risultati superiori alle membrane disponibili in commercio.
Dopo la preparazione istologica e l'acquisizione di immagini al microscopio ottico, l'analisi delle immagini eseguita mediante Photoshop e ImageJ ha fornito dati come la relazione tra volume osseo / volume totale; osso rigenerato; particelle residue di innesto; tessuto molle; indice di conducibilità; spessore trabecolare e spaziatura trabecolare; rapporto di contatto osso-impianto (BIC). Per l'analisi chimica e microscopica dei tessuti duri, SEM-EDS è stato utilizzato per la quantificazione del calcio e del fosforo in diverse regioni del campione osseo, dopo la lucidatura del preparato istologico e del suo rivestimento in carbonio (solo alcuni rapporti sono disponibili in letteratura). È stato analizzato un campione contenente un impianto in titanio per valutare la sua interfaccia con l'osso e l'osso corticale e l'osso spongioso cresciuti attorno all'impianto. Anche se i valori di riferimento mancano ancora in letteratura, è stato possibile rilevare diversi gradi di mineralizzazione all'interfaccia (diversa concentrazione di calcio) e tra osso spongioso e corticale.
Conclusioni
Le membrane prodotte mediante ELS hanno mostrato risultati promettenti in termini di (i) affidabilità della produzione (ii) biocompatibilità (ii) proprietà antimicrobiche (iii) durata in SBF che consente la programmazione di ulteriori esperimenti in vivo con le membrane ottenute. Parallelamente, l'arricchimento sperimentale delle membrane PRF, che mostra anche l'adsorbimento di nHAp, deve ancora essere perfezionato per ottenere una distribuzione omogenea di nHAp sulla superficie della membrana.
Introduction and thesis objectives
General introduction
Tissue regeneration and tissue engineering represent highly multidisciplinary fields that combine different approaches to restore, maintain or improve tissue and organ functions of the human body [1]. These fields arise from the combination of materials engineering, biology and medicine, and explore the use of biomaterials, cells, growth factors (GFs), nanomedicine, immunomodulation, gene therapy and other strategies. The design of biomaterials is aimed at the preparation of hydrogels, scaffolds, membranes, and fillers that have to allow, sustain and promote cell adhesion, migration, proliferation, differentiation and function.
bone regeneration (GBR) follows the principle of tissue isolation and separation with volume maintenance. GBR guides the bone matrix through the formation of mineralized bone, separating it from faster turn-over tissues such as connective and epithelial, and preserving the newly formed bone tissue from tissue shrinkage and muscular activity (e.g. tongue, cheeks). Commonly used materials for GBR are affected by mechanical, biocompatibility and resorbability limits, which can be overcome by different strategies such as nanotechnologies. In this field, nanofibers and especially electrospun nanofibers, are interesting for applications in the production of membranes based on both synthetic and natural polymers. Besides, despite autologous bone substitutes have shown historically the best performances in bone regeneration, they bring the inner limitation of a second surgical donor site together with patient’s augmented mobility. From 2000 however, the easy technique of generate hemoderivatives directly from the patient’s peripheral blood in outpatient procedure gain attention in scientific and clinical community. Second generation of hemoderivatives such as leucocyte derived PRF (L-PRF) have shown promotion of soft tissue healing thanks to the platelet growth factors (PDGF, VEGF, IGF, EGF, FGF, TGFβ-1) [7]. However, probably because of the documented limited in vitro durability of GFs, they appear not enhancing bone healing. Nonetheless scientific research is exploring the tunability of these implantable fully autologous scaffolds and membranes with biocompatible and osteoinductive materials.
Bone regeneration via electrospun nanofiber membranes
Nanofibers in Tissue Engineering (TE) represent an extremely attractive subgroup of biomaterials due to their unique intrinsic properties. Nanofiber-based membranes or scaffolds exhibit high surface-to-volume ratio, which allows an improved cell adhesion. Moreover, these structures offer tunability with proteins, drugs and ligands. The mechanical and morphological properties are even more promising thanks to the customizing dimensions of the fibers, their orientation and packing, porosity and density. Finally, the resulted three-dimensional structure of the obtained nanostructured material mimics the morphology of the extracellular matrix, consisting predominantly of collagen fibrils, coupled with elastin and other macromolecules such as glycoproteins [8]. Furthermore, nanofibers can promote specific cellular functions such as adhesion, proliferation, differentiation, and modulate stem cell behaviour [9,10].
prepare nanofibers. Electrospinning (ELS) technique has risen its popularity since its early development in the 1930s [15] along with the refinements of its basic components and setup. This technique is used for polymeric solutions that can be modified and enriched with bio-active molecules. Thanks to their features, electrospun nanofibers have been attractive also in the dental field: periodontal regeneration [16], coatings for caries prevention [17], enrichment of resin composites [18], implant surface modification [19], wound healing of mucosa [20], drug releasing systems [21] and bone regeneration [22]. In GBR, membranes should maintain the shape of the defect in which the bone is stimulated to regenerate. The membrane should also protect the initial blood clot from any compression, shielding the bone matrix during maturation from infiltration of soft tissues cells. Therefore, these membranes should maintain suitable mechanical properties at least for three months of permanence exhibiting at the same time a proper bio-degradability which avoids second surgery for patients [23,24].
aspect prospectively gives this basis research a clinical purpose and an easier translation from the bench to the bedside [44]. Moreover, among biodegradable synthetic polymers PCL is the one with the slowest degradation kinetic [45].
(Geistlich Pharma AG, Wolhusen, Switzerland); whereas the non-degradable benchmark product is a titanium reinforced expanded polytetrafluoroethylene (ePTFE) membrane named Cytoplast® (Osteogenics Biomedical, Lubbock, USA). Hence, is evident the paramount disparity in terms of technology and fabrication processes between the commercially available membranes and the state of the art of scientific research. At the time of writing this PhD Thesis, the evidences reported in literature remains at the in vitro or in vivo (animal model, small sizes) level. Thus, someone might argue if the promising results reported by the basic research, may be similarly good in the human application or even superior to the current outcomes.
Bone regeneration with enriched hemoderivatives
Platelet-rich fibrin (PRF) belongs to a second generation of platelet concentrates, as its processing is simplified and no biochemical treatment of blood is required [50]. Nowadays, it is known that platelets play a crucial role not only in hemostasis, but also in the wound healing process. In 1974 the theory concerning a possible regenerative potential of platelets was introduced for the first time; Ross et al. in 1974 [51] were the first authors to describe a platelet growth factor. Gassling et
al. [52] published a study in 2010 where the authors compared the use of the PRF membrane with
the collagen membrane (Bio-Gide) to evaluate which membrane was the best suitable form GBR. The authors observed that after the activation of the platelets trapped inside the fibrin matrix, the released growth factors stimulated the mitogenic response in the periosteum, inducing a better wound healing. Platelet-rich plasma (PRP) was used as a means to introduce growth factors such as PDGF, TGF-β and IGF-1 in the surgical site, in order to enrich the physiological clot to accelerate wound healing and stimulate bone regeneration [53]. Generally, a human blood clot consists of 95% of erythrocytes, 5% of platelets, less than 1% of leukocytes and millions of fibrin filaments. A PRP clot contains about 4% of erythrocytes, 95% of platelets and 1% of white blood cells [54].
anticoagulants, the platelets soon activate the coagulation cascade. The platelets are located in a massive way in the fibrin mesh, so as to release the growth factors. The fibrin network allows the progressive release of growth factors in a time frame of about 7-11 days. According to the study conducted by Simonpieri et al. [55], the use of the PRF membrane in bone grafting offers four advantages:
• First, the fibrin clot has an important mechanical role, in fact the PRF membrane maintains and protects the additional grafted biomaterials, as well as acting as a biological connector between the wound and the adjacent areas.
• Secondly, the PRF membrane facilitates cell migration, in particular for the endothelial cells necessary for neo-angiogenesis, vascularization and graft survival.
• Platelet cytokines (PDGF, TGF-β, IGF-1) are gradually released as the PRF matrix is reabsorbed, thus supporting a perpetual healing process.
• The presence of leukocytes and cytokines in the fibrin clot plays a significant role in the self-regulation of inflammatory and infectious phenomena within the grafted material.
In the literature there are numerous scientific articles which confirm that in the PRF protocol the absence of the use of thrombin and calcium chloride (as in PRP production) allows to obtain a fibrin clot with a physiological and very elastic molecular structure [56][57][50]. Furthermore, the growth factors being released for over a week, allow for rapid healing of soft tissues, particularly during the first two weeks of healing, which appear to be the most critical [58].
Conversely, enhanced bone regeneration remains still controversial. Apparently, based on scientific clinical evidence, PRF seems not to help to ameliorate bone regeneration efficiency. However there is a great effort in basic research to enrich the hemoderivatives with bone promoting agents such as alendronate [59], DBBM [60], β-TCP [61] with promising results compared with the bone promoting agents or PRF alone.
Bone regeneration analysis
high contrast. Bone biopsy is the way to harvest the specimen from the living body. The sample must be taken through the use of a drill with a low number of revolutions and under copious irrigation to avoid possible alterations to the bone structure. The analysis is based on the evaluation of three types of measurements that include perimeters, areas and lengths that together will give a presentation of the two-dimensional structure of the sample, highlighting areas, distances and distributions. What histomorphometric analysis usually aims at evaluating concerns neoformation and bone resorption, the distribution of the different components and the cell population. This aspect is fundamental for the analysis of regenerated bone starting from a scaffold or a membrane: parameters such as connectivity index, highlights the biocompatibility and osteoconductivity of a bone substitute. Taking into account, for example, the value of Tb.N, which indicates the number of trabeculae present, it can be referred to the mechanical properties of the analyzed bone.
The parameters related to bone microarchitecture, in fact, are particularly important for assessing bone quality, as they seem to be the main risk factor for fracture. This can be demonstrated through the Euler principle according to which an important resistance to compression is lost in the case in which a single horizontal trabecula is reabsorbed.
Histomorphometry represents the gold standard in the analysis of the two-dimensional structure of the bone, with particular attention to microarchitecture and the distribution of trabeculae. However, other techniques have been introduced as supplementary analysis, such as micro-CT, which provides a more detailed three-dimensional evaluation.
AIM of the research
Nanofiber based membranes and scaffolds represent an extremely interesting starting point for development of implantable device for bone regeneration purposes. Nowadays clinical practice is pretty far from the results and the prospective of the basic research, thus the implementation of new research protocols based both on electrospun synthetic and natural polymers together with the enrichment of autologous hemoderivatives represent the starting point of this thesis. Finally, the analysis of the efficacy of these devices need to be thoroughly investigated through qualitative and quantitative methods such as histomorphometry and SEM-EDS ultramolecular analysis of mineralized tissues. Thus, this thesis has three main objectives, as depicted in Fig. 1:
▪ the production of a novel nanostructured electrospun membrane based on PCL enriched with nAg as antimicrobial agent and CTL as bioactive molecule;
▪ the enrichment of a nanostructured autologous hemoderivatives membranes with HAp; ▪ the optimization and introduction as standard research protocol at University of Trieste of
histomorphometric analysis and SEM-EDS experimental application to the mineralized tissues analysis.
Fig. 1: Schematic representation of the thesis organization.
BIORESORBABLE ENGINEERED MEMBRANES FOR GUIDED BONE
REGENERATION WITH ANTIMICROBIAL PROPERTIES
CHAPTER I
Electrospun PCL bioactive membranes with antimicrobial properties
CHAPTER II
Autologous PRF membranes enriched with nHAp
CHAPTER III
Chapter 1 electrospinning in bone regeneration
Variables affecting electrospinning process
Despite the broad spectrum of polymers that can be electrospun, an equilibrium between the physical and chemical properties and the ratio between the solute and the solvents has to be thoroughly seek, along with the multiple variables that may affect the final morphology of the obtained fibers. A list of the relevant parameters is provided in the following table (Tab. 1) [63–65].
Solution parameters Process parameters Environmental parameters
Viscosity Voltage Humidity
Concentration Flow rate Temperature
Conductivity Shape of collector Air flow of the chemical hood Dielectric constant Needle gauge -
Surface tension Distance -
Charge of jet Angle -
Solvent type Motion -
Polymer type - - Polymer molecular weight - - Polymer solubility - - Boiling point - -
Tab. 1: a list of the relevant parameters that may affect the final morphology of electrospun
nanofibers is provided.
Molecular weight depends on the chain length of the polymer and can be related to the entanglements of the molecules. This fact explains why high molecular weight results in viscous solutions compared to low molecular weight. Therefore, the molecular weight of the polymer should be correctly considered for the selection of solvents and concentrations. Indeed, if the solution exhibit too high viscosity, this will hamper the flow through the capillary and the polymer may dry up or drip at the needle tip. Conversely, solutions with relatively low concentration will result into droplets.
Solubility and boiling point of the solvent are paramount factors. Volatile solvents are ideal options due to rapid evaporation during the transit from the needle tip to the collector [66]. High boiling points solvents may not evaporate completely prior to hit the target, resulting in flat ribbon shape fibers (Fig. 2) instead of circular fibers, presence of beads or other defects (Fig. 3) [67]. The volatility of the solvent may affect the final microscopic characteristics of the obtained fibers including porosity, shape and size.
Fig. 2: nanofiber-based membrane obtained with the following parameters: chitosan 2.5% w/V +
Fig. 3: nanofiber-based membrane obtained with the following parameters: PCL 6% w/V in
DCM/methanol (MeOH) 7:3, 17 KV of potential, 27G needle, flow rate of 0.6 ml/h. The formation of multiple beads can be appreciated. Quanta250 SEM, FEI, Oregon, USA; 2000x.
proper regulation of the flow rate is also function of the distance between the metallic needle and the collector. This parameter should allow for the correct solvent evaporation during the transit between the source and the target.
Depending on the polymer and the solvent, the needle diameter can vary. Smaller internal diameter reduces the probability of occlusion of the spinneret due to less exposure time of the jet to the environment. Reduction in needle internal diameter increases the surface tension of the solution corresponding to smaller droplet. This causes the decrease of Jet acceleration. So jet gets more flight time before deposition; this results in smaller diameter of the fibers [65]. Usual needle diameters are reported to be from 18G to 30G [72–74].
Apart from solution and processing parameters, also the ambient parameters (e.g. humidity, pressure and temperature) influence the fiber morphology. The variation in humidity while spinning polystyrene solutions was studied and it was showed that an increase in the humidity results in the appearance of small circular pores on the surface of the fibers; a further increase in the humidity leads to the pores coalescing [87]. In 2004 the effect of temperature, ranging from 25 to 60 °C, was investigated on the ELS of polyamide-6 fibers and it was found that higher temperatures led to smaller fiber diameter. The authors attributed this phenomenon to the decrease in the viscosity of the polymer solutions at increased temperatures [86].
Material and methods
Fig. 4: experimental set-up
Literature review
Material and tools acquisition
ELS device assembling
Experimental combination of solution and solvent parameters
Nanostructured PCL membrane production
Plasma air activation
Enrichment with CTL + nAg
Phisical and chemical analysis
Morphological analysis
Mechanical analysis
Biological and microbiological analysis
Contact angle analysis of the activated membranes
Electrospinning device assembly
This technique was firstly applied in 1934 by Anton Formhals and represents a combination of two techniques which are the electrospray and the spinning of fibers [88]. A high electric field is applied to both the syringe needle, which contains a polymeric solution, and to the collector (Fig. 5).
Fig. 5: schematic representation of the essential set-up of an electrospinning device.
The collector and the syringe needle are kept at the proper distance one from the other. Metallic plates, aluminium foils and rotating drums can be used as target for the collection of nanofibers during the electrospinning process. The potential difference is able hence to overcome the surface tension of the polymeric solution ejected from the needle tip and assume the so called “Taylor cone” configuration [89] (Fig. 6). This process shapes the polymeric solution into a jet of charged fluid that is electrostatically attracted by the collector. The solvent evaporates during this transit from the needle to the collector allowing for the accumulation of dry fibers on it.
Fig. 6: Taylor cone obtained with the following parameters: a solution of polycaprolactone (PCL)
Experimental setup
The set-up for carrying out the ELS experiments (Fig. 5) consists of several components: first, a syringe containing the polymeric solution, a needle is applied to it and the syringe is positioned on a syringe pump, which has the function of regulating the flow rate with which the polymer solution will be extruded from the syringe; in front of this system there is a target covered with an aluminium sheet on which the fibers will be collected during the formation phase. The needle positioned on the syringe and the collector are both connected to an electric potential generator to allow the formation of an electric field at the two ends capable of breaking the surface tension of the solution containing the polymer dissolved in the solvents and thus generating the fiber that will deposit on the collector.
Fig. 7: image of the set-up used during the PCL membrane production experiments. The syringe
pump is positioned under chemical hood with regulated air-flow.
Materials
(Holliston, MA, USA). Silver nitrate (AgNO3), ascorbic acid (C6H8O6), LDH (lactate dehydrogenase)-based TOX-7 kit, in-vitro toxicology assay (Resazurin dehydrogenase)-based, Alamar Blue) TOX-8 kit, phosphate buffered saline (PBS), Luria–Bertani (LB) broth, LB Agar and Brain Heart Infusion (BHI) were purchased from Sigma-Aldrich (Chemical Co. USA). Trypsin/EDTA solutions, Fetal Bovine Serum (FBS), penicillin streptomycin 100X, l-glutamine 100X and Dulbecco’s modified Eagle’s medium (DMEM) were purchased from EuroClone (Milan, Italy). All other chemicals were of analytical grade.
Preparation of PCL solutions
To obtain the polymeric solutions, PCL was solubilized by overnight stirring, in a first step in one of the solvents: THF or DCM. In the second solubilization phase, an additional solvent (DMF) was added. The mixture thus obtained was stirred for at least 30 minutes, in order to allow the mixing of the two solvents.
The target is positioned on a slit which allows the experimenter to adjust the distance between it and the syringe needle. With the same mixture of solvents used and the polymer concentration, the following parameters have been changed: voltage (kV), needle-target distance (cm), flow rate (mL/h) and internal diameter of the needle (G); furthermore, fiber collection time vary from a few seconds (15 s) up to hours, depending on the desired width of the membrane.
Solvents and parameters of PCL solutions for ELS
Once the solutions were obtained, these were loaded into a glass syringe (inner diameter 9 mm), to which a metal needle was applied, whose internal diameter varied according to the desired product. Subsequently, the syringe containing the polymeric solution was placed on the syringe pump, the needle and the target (covered by an aluminium sheet) were connected to the generator of electric potential and, finally, after the ignition of the generator and syringe pump, the procedure was activated. The parameters that have been changed in this experimental phase are reported in the following tables and are:
o Polymer concentration (%) o Ratio of the solvents used
o Distance between the needle and the fiber collector (cm) o Electric field intensity (kV)
o Internal diameter of the needle (G) o Flow rate (mL/h)
Each parameters range was selected after a thorough analysis of the recent literature, and varied in order to get the most reproducible nanofiber-based membrane with the fiber diameter in the nanoscale, homogeneity of diameter dimension distribution and pores size, the absence of impurities and defects. The rationale of the parameters combination always arose from the information published in the literature. All these efforts were made to obtain a stable and easy to handle membrane for surgical purposes.
THF/DMF
PCL concentration (%) Solvent ratio Needle-collector distance (cm)
Voltage (kV) Needle gauge
(G) Flow rate (mL/h) TELS (min.) 6 1:1 3:7 7:3 12 15 24 15 20 27 23 25 27 0.4 0.6 0.8 1.0 1.5 2.0 5 8 1:1 15 24 15 27 1 5
Tab. 2: the table shows the variation of the experimental parameters performed during the
experimental session, aimed at investigating the THF / DMF solvent mixture.
DCM/DMF
PCL concentration (%)
Solvent ratio
Needle-collector distance (cm)
Voltage (kV) Needle gauge
(G) Flow rate (mL/h) TELS (min.) 6 7:3 12 24 10 23 25 0.6 1.0 5 8 7:3 24 10 17 27 0.6 5 10 7:3 24 10 17 27 0.6 5 12 7:3 8 16 24 10 15 17 24 23 25 27 0.2 0.4 0.6 1.0 5 30 60
Tab. 3: the table shows the variation of the experimental parameters performed during the
Scanning electron microscope (SEM)
The central part of the obtained samples has been harvested (1 cm2). This portion was deposited on a carbon double-sided tape, previously placed on an aluminium sample holder (stub). Subsequently, the metallization with gold of the samples was performed using a Sputter Coater K550X (Emitech, Quorum Technologies Ltd, UK). Samples were analyzed with a scanning electron microscope (Quanta 250 SEM, FEI, Oregon, USA). The micrographs were obtained using the secondary electron detector. The working distance was set at 10 mm, to obtain the appropriate magnifications the acceleration voltage was set at 30 kV. At least 10 images were taken for each sample, at different magnifications and in different portions of the sample. The dimensional analysis of the fibers was performed with the help of a plugin, DiameterJ, created for the ImageJ analysis software; DiameterJ is able to analyze an image and calculate the diameter of the nano-microfibers for each pixel present along the fiber axis and produce, as a final result, a series of quantitative data. The average of the fiber diameters and the respective standard deviations obtained with DiameterJ were subsequently processed using Excel 2010 software.
Plasma-air treatment
Thanks to the collaboration with Dr. Denis Scaini of the International School for Advanced Studies of Trieste (SISSA), the PCL membranes were subjected to a plasma-air treatment, in order to increase the hydrophilicity of this polymer. This operation was performed by setting the plasma air at low (6.8 W) and medium (10.5 W) power, for a duration of 5 minutes, using the PDC-32G Plasma Cleaner (Harrick Plasma, Ithaca NY, USA), RF frequency 8-12 MHz.
Contact angle
of membrane: plasma-air-activated PCL membranes, low-power activated plasma-air membranes, activated and CTL-coated membranes and activated membranes coated with CTL-nAg.
CTL production
CTL (lactose modified chitosan) was prepared according to the procedure reported elsewhere starting from highly deacetylated chitosan [90]. The composition of CTL was determined by means of 1H-NMR and resulted to be: glucosamine residue 27 %, N-acetylglucosamine 18 % and 2-(lactit-1-yl)-glucosamine 55 %. The calculated relative MW of CTL is around 1.5 × 106.
CTL nAg production
Silver nanoparticles (nAg) were obtained by reducing silver ions with ascorbic acid in CTL solution. Freeze-dried CTL was dissolved in deionized water to obtain a 4 g/L solution. Silver nitrate (AgNO3) was added to CTL at final concentration of 1 mM; then, ascorbic acid was added at final concentrations of 0.5 mM. The solution was kept for 4 hours at room temperature in darkness and then stored at 4 °C.
Computerized micro-tomography (μCT)
In order to evaluate the thickness of the PCL membranes obtained after 30 min and 60 min of fiber deposition on the collector, computed micro-tomographies were obtained by means of a customized cone-beam system called TOMOLAB. The samples were placed on the rotor and the acquisitions were performed with the following parameters: source-sample distance (FSC), 80 mm; distance source-detector (FDS), 250 mm; magnification, 3.1x; binning, 2x2; resolution, 8 μm; dimension of tomographic projections (pixels), 1984x1984; number of tomographic projections, 1440; number of slices, 1332; energy of beam, E = 40 kV; beam intensity, I = 200 μA; exposure time, 1.3 s. The process of reconstruction of the slices and correction of the artefacts deriving from the acquisition of the projections were carried out with the Cobra Exxim software. The Fiji software [91] was used for analysis of scans and measurement of membrane thicknesses.
CTL confocal microscope analysis of adsorption on electrospun PCL membranes
discs, obtained with the help of a circular scalpel (punch) for 6 mm diameter skin biopsies. On each disk a volume of 200 μL of fluorescent CTL (CTL-FITC) was added and allowed to adsorb overnight. The following day, the samples were washed 3 times in distilled water (5 minutes for each wash) and then allowed to dry under a biological hood.
The images of the samples on which the CTL-FITC was adsorbed were acquired with a Nikon Eclipse C1 microscope, with a Nikon Plan Fluor lens (numerical aperture: 2.1, dry) using an argon laser (488 nm) and an acquisition channel at 515-530 nm. The images obtained were analyzed using Fiji software.
FTIR analysis
The polycaprolactone, the CTL, the PCL membranes (obtained by applying a 10 kV and 17 kV voltage) treated with plasma air at low and medium power, on which the CTL was adsorbed were characterized by infrared spectroscopy in transform of Fourier (Nicol ATR model 6700 Thermo Scientific, MI, Italy). The infrared spectrum of the samples was measured in a wave number range of 4000–400 cm-1. All spectra were obtained through the accumulation of 32 scans with a resolution of 4 cm-1.
Raman Analysis
The Raman spectra were obtained using a B & WTek iRaman plus spectrometer, with a 20x objective. A 785 nm laser, whose power was 300 mW, was used as excitation source. For each spectrum 3 accumulations have been acquired, each lasting 5 seconds. The spectra have been corrected and normalized with the baseline.
Electrothermal atomic absorption spectrometry (ETAAS)
The statistical analysis was performed with the Origin software, verifying that the data followed a normal distribution (Kolmogorov Smirnov test) and that the variances were homogeneous (Levene’s test). The data were compared with each other with an ANOVA t-test applying Tukey's post-hoc correction.
Simulated body fluid (SBF) production and membrane dissolution assay
The SBF was prepared by dissolving reagent-grade chemicals of NaCl, NaHCO3, KCl, K2HPO4, MgCl2·6H2O, CaCl2, Na2SO4 in distilled water and buffering at a pH of 7.40 with tris(hydroxymethyl)aminomethane (CH2OH)3CNH2 and 1.0 M HCl at 36.5 °C. The final ion concentrations is nearly equal to those of human blood plasma (Na+ 142.0, K+ 5.0, Mg2+ 1.5, Ca2+ 2.5, Cl− 147.8, HCO
3− 4.2, HPO42− 1.0, SO42− 0.5 mM). Weight variation of membranes was evaluated after 1, 2, 3, 4, 8, 12, 16, 20 weeks of soaking in SBF at 37 °C.
Mechanical testing with DMA
Mechanical resistance of the membrane was tested with uniaxial tensile tests by means of a Dynamic Mechanical Analysis (Electroforce 3300, TA Instruments, 1 59 Lukens Dr, New Castle, DE 19720, US). For the purpose the same membranes used in the dissolution assay, have been cropped in a dog-bone shape thanks to a custom templated produced with a 3D printer (Ultimaker3 Extended). The testing machine mounted a load cell of 22 N with a maximum excursion of deformation of 25 mm. Samples were tested at the constant deformation of 1 mm/min. Maximum strength and strain, or strain and stress at break, and elastic modulus were recorded.
Cell culture and in vitro tests
The line of human osteoblastic cells derived from osteosarcoma (MG63, ATCC number: CRL-1427) was grown in DMEM supplemented with 10% FBS, penicillin 100 U/mL, streptomycin 0.1 mg/mL, L-glutamine 2 mM, at 37° C and with 5% pCO2. Cells were passed (trypsin 0.25%) twice a week when the confluence level was estimated at about 70-90% of the available culture space.
SEM and cell adhesion and proliferation tests
For cell adhesion and proliferation experiments, 3000 MG63 cells (ATCC number: CRL-1427) were seeded on each sample in 40 μL of DMEM medium supplemented with 10% FBS, penicillin 100 U/mL, streptomycin 0.1 mg/mL, L-glutamine 2 mM, at 37° C and with pCO2 5%. After 4 hours (to ensure cell adhesion), 200 μL of complete DMEM medium were added to each well. For the SEM analysis 2 disks were prepared for each experimental time and for each type of membrane. At each experimental time interval, the membranes were washed in PBS and fixed for 45 minutes with 4% PFA in PBS. Subsequently, the samples were washed in deionized water and dehydrated with 20-minute steps in ethanol solutions in water with increasing concentration (30%, 50%, 70%, 90%, 100%) and subsequently in solutions of hexamethyldisilazane in ethanol (50% and 100%). After the evaporation of hexamethyldisilazane, the membranes were metallized with gold (Sputter Coater K550X, Emitech, Quorum Technologies Ltd, UK) and analyzed by SEM (Quanta250 SEM, FEI, Oregon, USA) using the secondary electron detector and using an acceleration voltage of 30 kV.
Cell proliferation
Cell proliferation was tested by seeding MG63 cells on membrane surface and analysing their proliferation rate using Alamar BlueTM test at different time points.
For the Alamar BlueTM test 8 disks were prepared for each type of membrane (treated with low-power plasma-air, treated with plasma-air and CTL adsorbed and treated with plasma-air and adsorbed CTL-nAg), of which 2 are been used as white; the samples were placed in a 96-well flat-bottomed culture plate. In each well containing the membrane, 3000 MG63 cells were placed and left for 24 h in a humid atmosphere of 5% CO2 at 37° C. Adhesion and cell growth were assessed by Alamar BlueTM assay. The tests were performed at 1, 3, 6 and 8 days. At each experimental time point the culture medium was removed and 200 μL of Alamar BlueTM were added to DMEM (10%) for each well. After 4 hours of incubation at 37° C in the dark, 150 μL were taken from each well for fluorescence reading; each well was washed with PBS and 200 μL of medium were subsequently added. Fluorescence reading was performed using a GloMax Multi + Detection System (Promega) spectrofluorimeter using an excitation wavelength of 525 nm and collecting the fluorescence emission in the range 580-640 nm.
Cell toxicity
Cell toxicity was tested measuring the lactate dehydrogenase (LDH) released from MG63 cells cultured in the presence of the membranes and in the presence of negative and positive controls of toxicity. In vitro cytotoxicity of PCL-CTL-nAg was evaluated by using lactate dehydrogenase cytotoxicity assay (SIGMA TOX-7LDH assay). UV-sterilized membranes were placed in Dulbecco’s modified Eagle’s medium, inactivated fetal bovine serum 10 %, penicillin 100 U/mL, streptomycin 100 μg/mL and L-glutamine 2 mM for 24 h. After 24 h of incubation, the cytotoxicity test was performed by direct contact of the cells with the swollen membranes (20 mg per well). Cells were seeded into 24-well plates (30000 cells per well) and incubated 24 hours before the cytotoxicity test. The experiments were performed in triplicate. Cells were than incubated for 24 and 72 hours with membranes. After 24 and 72 hours, the medium was collected and the test was performed following the manufacturer’s protocol. The absorbance was measured at 490 nm and 690 nm, with a Tecan Nano Quant Infinite M200 Pro plate reader. The cytotoxicity was calculated using the following equation:
normalizing the values for the total LDH of the control cell lysate.
Polystyrene (PS) was used as a negative control; zinc embedded polyurethane (PU/Zn) membrane and Triton X-100 0.01% in PBS were used as positive control.
Biofilm inhibition
The antibacterial activity of PCL membranes containing silver nanoparticles was tested in terms of biofilm inhibition on Pseudomonas aeruginosa (ATCC 27853) and Staphylococcus aureus ATCC 25923, using an MTT test to assess biofilm viability.
membranes covered by CTL-nAg; all the membranes used were not deprived of the aluminium sheet used in the production phase by ELS. As a white of bacterial growth, test tubes containing only 10% LB medium were used, without the bacterial strains.
Cells from two different bacterial strains, Pseudomonas aeruginosa ATCC 27853 and Staphylococcus
aureus ATCC 25923 were grown on an LB agar plate and subsequently allowed to grow overnight in
Results
PCL-based solutions
Two formulations of solvent mixtures were used for the preparation of electrospun PCL membranes. Specifically, Tetrahydrofuran (THF) with N, N-dimethylformamide (DMF) was used based on the work of Nhi et al. [92] and dichloromethane (DCM) with DMF referring to the article by Du et al. [93].
Here, various parameters have been changed in order to optimize the experimental set-up for the best production of electrospun polymeric membranes. Table 4 summarizes the results of the parameter optimization phase.
Solvents PCL Concentration (%) Fibers mean diameters (nm) Observations THF/DMF 6 150-275 Beads
8 200-250 Aggregates, instable solvent
DCM/DMF 6; 8; 10 350-700 Beads and aggregates
12 500-1900 No defects, homogeneous fiber diameter
Tab. 4: Sum of the results obtained with different parameters of ELS.
Fig. 8: Images obtained by SEM analysis of PCL nanofibers in 6% THF / DMF 7: 3 with set-up of 24
cm of needle-collector distance, 15 kV, needle 25 G in which the flow was varied rate: 1) 1 mL / h; 2) 1.5 mL / h; 3) 2 mL / h (scale bar: 50 μm).
Fig. 9: The average diameters (μm) and the standard deviations of PCL fibers at 6% in THF / DMF 7:
3 are reported, obtained with the following set-up: PCL 6% in THF / DMF 7: 3 with set-up of 24 cm away needle-collector, 15 kV, needle 25 G in which the flow rate has been changed: 1) 1 mL / h; 2) 1.5 mL / h; 3) 2 mL / h
In the attempt to obtain nanofibers without defects and to decrease the average fiber diameter, in the subsequent tests the PCL concentration parameters were changed (8% w / v), the ratio between the two solvents (1:1 ; 3:7 and 7:3), the distance between needle and collector (15 and 24 cm), the flow rate (0.4; 0.6; 0.8; 1; 1.5; 2 mL/h), the voltage (15 kV; 20 kV and 27 kV) and needles used (23 G; 25 G and 27 G). Despite the variation of all the parameters indicated above and the obtainment of nanometric fibers, the obtained membranes also showed numerous beads and defects, probably for the instability of THF used in the realization of such membranes. For this reason, it was considered appropriate to use other types of solvents, in order to obtain a reliable system that is as free of defects as possible.
Fig. 10: image obtained by SEM analysis in which PCL fibers are reported at 15% w / v,
needle-collector distance 24 cm, voltage 27 kV, needle 21 G and flow rate 10 mL / h (scale bar: 30 μm).
Fig. 11: images obtained by SEM analysis showing the PCL fibers at 12% w / v, needle-collector
distance 24 cm, voltage A) 10 kV and B) 17 kV, needle 27 G and flow 0.6 mL / h (scale bar: 50 μm).
Sol.A_1 7kV Sol.A_1 0kV Sol.B_1 7kV Sol.B_1 0kV Sol.C_1 7kV Sol.C_1 0kV 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 Diamet ro me dio fibr e ( m) Soluzione A_17 kV Soluzione A_10 kV Soluzione B_17 kV Soluzione B_10 kV Soluzione C_17 kV Soluzione C_10 kV
Fig. 12: Histogram showing the average diameters of the three PCL solutions 12% w / v in DCM /
DMF in a 7: 3 ratio, needle-collector distance of 24 cm, flow 0.6 mL / h and needle 27G, in which 17 kV and 10 kV are the potentials that have been changed.
μCT examination of PCL membranes
The images obtained by computerized micro-tomography (μCT) and subsequent data processing were carried out with the aim of determining the thickness of the membranes. This information is in fact indispensable especially in view of the mechanical characterization of these membranes. PCL membranes obtained by applying a potential of 10 kV and left to deposit on the collector for 30 minutes were examined, 17 kV electrospun membranes for 30 minutes and membranes obtained
by applying a 17 kV potential and made to deposit on the target for 60 minutes; the differences in terms of thickness between the aforementioned samples are noticeable in Figure 13.
Fig. 13: Image obtained from the investigation using µCT. A) 10 kV electrospun for 30 min, from B)
17 kV deposited for 30 minutes and from C) 17 kV obtained after 60 minutes of deposition (scale bar: 100 µm). The processing of data from the μCT allowed us to quantify the thickness of the membranes (Table 3).
Thickness (μm)
10 kV 30 min 46.6 ± 3.5 17 kV 30 min 137.3 ± 7.0 17 kv 60 min 215.6 ± 22.1
Tab. 3: average thicknesses of the PCL membranes at 12% w / v, 24 cm of needle-collector distance,
0.6 mL / h, 27G needle and 10 kV potential (electrospun for 30 minutes) and 17 kV (electrospun for 30 min and 60 min), as obtained from μCT analysis.
Mechanical testing stability of electrospun membranes
Membrane stability was evaluated as weight variation of samples soaked in SBF. The results, reported in Fig. 14, show that membranes are stable over time in terms of weight loss, indeed the slight increase of weight that can be observed is constant and can be ascribed to a retention of liquid after the evaporation.
Fig. 14: weight variation graph has been elaborated for both pristine and aged membranes.
Fig. 15: variation of the membranes Young modulus over time. A slight decrease of the average
value can be appreciated.
Fig. 16: Variations of the maximum strain over time. A constant behavior can be appreciated
during aging until 12 weeks.
As observed for the maximum strain, the maximum stress in almost constant for 12 weeks; the variability of the values up to 12 weeks can be related, as for the Young modulus, to the variability of the electrospinning process.
Fig. 17: Variations of the maximum stress over time. A constant behavior can be appreciated
during aging until 12 weeks.
All together, these results show that (Fig. 15 and 17) the membranes are stable even after weeks of immersion in SBF. It should be noted that the elasticity slightly decrease over time and that the parameters of maximum stress and maximum deformation change as the membranes break after 16 weeks. This behavior of PCL nanostructured membranes satisfy the mechanical resistance desirable for GBR purposes [47].
CTL e CTL-nAg adsorption on PCL membranes
In the series of experiments and characterization of polymeric fibers, it was possible to evaluate the adsorption of CTL on PCL fibers and how this influences the morphology of the fibers themselves. Furthermore, the effect of low and medium power plasma-air treatment on fiber morphology was evaluated by depositing the samples both in distilled water and in two CTL 2 mg/mL solutions having different pH (pH = 4.5 and pH = 7). The analyzed experimental conditions are reported as follows: membranes prepared at 10 and 17 kV were tested, on which CTL was adsorbed at pH 4.5 or 7, after treatment with low or medium power plasma. Controls were analyzed membranes not treated with plasma or with CTL and membranes treated with plasma and distilled water.
No morphological variation of the fibers was observed by qualitatively comparing the images obtained with the SEM of the membranes produced at the two different voltages (10 kV and 17 kV) not treated with plasma-air and immersed in water or in a CTL solution.
The qualitative considerations were confirmed by the quantitative analysis of the average diameters of the fibers in which no significant variation was shown (Fig. 18).
Fig. 18: representative image of a 17 kV electrospun membrane with CTL adsorbed on PCL fibers
(left) (scale bar: 50 μm) and histogram showing the average diameters of the PCL at 12% p / v not-treated with plasma air (10kV and 17kV), immersed overnight in water or in a CTL solution (right).
The plasma-air treatment at low and medium power does not alter either the morphology or the average diameter of the PCL fibers (Fig. 19).
10 kV N T 10 kV b assa p ot 10 kV m edia po t 17 kV N T 17 kV b assa p ot 17 kV m edia po t 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 10 kV Non trattate 10 kV bassa potenza plasma 10 kV media potenza plasma 17 kV Non Trattate 17 kV bassa potenza plasma 17 kV media potenza plasma
Diamet ro me dio fibr e ( m)
Fig. 19: representative image of PCL membrane 12% p / v electro-wired at 17 kV to which a
medium-power plasma-air treatment (left) (scale bar: 50 μm) was applied and histogram showing the
M ea n f iber d ia met er ( µ m) 10 kV non-treated 17 kV non-treated 10 kV non-treated + CTL 17 kV non-treated + CTL 10 kV N T 10 kV b assa p ot 10 kV m edia po t 17 kV N T 17 kV b assa p ot 17 kV m edia po t 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 10 kV Non trattate 10 kV bassa potenza plasma 10 kV media potenza plasma 17 kV Non Trattate 17 kV bassa potenza plasma 17 kV media potenza plasma
Diamet ro me dio fibr e ( m) 10 kV non-treated 10 kV Low energy plasma 10 kV medium energy plasma 17 kV non-treated 17 kV Low energy plasma 17 kV medium energy plasma
average diameters of the PCL membranes at 12% w / v (10 kV and 17 kV) treated with plasma-air at low and medium power (right).
The PCL 12% w / v membranes (10 kV and 17 kV) treated with low-power plasma-air and placed in water (without CTL adsorption) do not undergo alterations, while the two types of membranes treated with plasma-air at medium power exhibit an altered morphology, in which the fibers are merged with each other; this could be explained by the fact that the medium power treatment weakens the structure of the fibers, which, being more hydrophilic, tend to absorb water and modify accordingly. This behavior could be exploited for the encapsulation of molecules and hydrophilic drugs. Despite the morphological variation of the fibers, the average of the fiber diameters does not show significant changes (Fig. 20).
10kV b assa p ot+H2O 10kV m edia po t+H2O 17kV b assa p ot+H2O 17kV m edia po t+H2O 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6
10 kV bassa potenza plasma + H2O 10 kV media potenza plasma + H2O 17 kV bassa potenza plasma + H2O 17 kV media potenza plasma + H2O
Diamet ro me dio fibr e ( m)
Fig. 20: representative image of PCL membrane 12% p / v electrospun at 17 kV and placed in distilled
water to which a medium-power plasma-air treatment (left) (scale bar: 50 μm) and histogram was applied showing the average diameters of PCL membranes at 12% w / v (10 kV and 17 kV) treated with plasma air at low and medium power and placed in distilled water (right).
Finally, the samples obtained were compared with the two different potential intensities (10 kV and 17 kV), treated with low-medium power plasma-air on which the CTL was adsorbed. Both from the qualitative analysis using scanning electron microscopy (SEM) and from the subsequent processing of the data it was verified that the addition of the CTL on these fibers does not alter either the morphology or the average thickness of the diameters (Fig.21).
10kV b assa p ot+H2O 10kV m edia po t+H2O 17kV b assa p ot+H2O 17kV m edia po t+H2O 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6
10 kV bassa potenza plasma + H2O 10 kV media potenza plasma + H2O 17 kV bassa potenza plasma + H2O 17 kV media potenza plasma + H2O
Diamet ro me dio fibr e ( m) M ea n fi be r di am eter ( µ m)
10 kV Low energy plasma + H2O
10 kV medium energy plasma + H2O
17 kV Low energy plasma + H2O
